US11503994B2 - Detection of pathologies in ocular images - Google Patents

Detection of pathologies in ocular images Download PDF

Info

Publication number
US11503994B2
US11503994B2 US16/721,902 US201916721902A US11503994B2 US 11503994 B2 US11503994 B2 US 11503994B2 US 201916721902 A US201916721902 A US 201916721902A US 11503994 B2 US11503994 B2 US 11503994B2
Authority
US
United States
Prior art keywords
eye
image
measurement
pathology
image data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US16/721,902
Other languages
English (en)
Other versions
US20200196853A1 (en
Inventor
Jano van Hemert
Derek Swan
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Optos PLC
Original Assignee
Optos PLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Optos PLC filed Critical Optos PLC
Publication of US20200196853A1 publication Critical patent/US20200196853A1/en
Priority to US17/942,132 priority Critical patent/US12042225B2/en
Application granted granted Critical
Publication of US11503994B2 publication Critical patent/US11503994B2/en
Priority to US18/748,540 priority patent/US20240341589A1/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
    • A61B3/102—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for optical coherence tomography [OCT]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/0016—Operational features thereof
    • A61B3/0025—Operational features thereof characterised by electronic signal processing, e.g. eye models
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/02—Subjective types, i.e. testing apparatus requiring the active assistance of the patient
    • A61B3/06—Subjective types, i.e. testing apparatus requiring the active assistance of the patient for testing light sensitivity, e.g. adaptation; for testing colour vision
    • A61B3/063—Subjective types, i.e. testing apparatus requiring the active assistance of the patient for testing light sensitivity, e.g. adaptation; for testing colour vision for testing light sensitivity, i.e. adaptation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
    • A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
    • A61B3/14—Arrangements specially adapted for eye photography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/48—Other medical applications
    • A61B5/4842—Monitoring progression or stage of a disease
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/18—Eye characteristics, e.g. of the iris
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10101—Optical tomography; Optical coherence tomography [OCT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30041—Eye; Retina; Ophthalmic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images

Definitions

  • Example aspects herein generally relate to the field of image processing and, more particularly, to the processing of an ocular image to facilitate identification of the presence and/or location of a pathology in an imaged portion of an eye.
  • An ocular image of a portion of a subject's eye acquired by an ocular imaging system may have one or more regions containing some indication of a pathology that would not be present in a comparable image of a healthy eye, and the presence of such region(s) may be indicative of a disease or an otherwise unhealthy state of the subject.
  • a fluorescein angiogram of the subject's retina may contain one or more regions where blood vessel leakage appears to have occurred, which can be a sign of macular oedema.
  • Ocular images require an experienced ophthalmologist to acquire and assess. However, existing techniques are time-consuming for the ophthalmologist and/or the patient, and the assessment performed may be prone to error.
  • the present inventors have devised, in accordance with a first example aspect herein, a computer-implemented method of searching for a region indicative of a pathology in an image of a portion of an eye acquired by an ocular imaging system.
  • the method comprises receiving image data defining the image, and searching for the region in the image by processing the received image data using a learning algorithm trained on image data defining images of the portion of healthy eyes, and image data defining images of the portion of unhealthy eyes each having at least one region that is indicative of the pathology.
  • a location of the region in the image is determined, an instruction is generated for an eye measurement apparatus to perform a measurement on the portion of the eye to generate measurement data, using a reference point based on the determined location for setting a location of the measurement on the portion of the eye, and the measurement data is received from the eye measurement apparatus.
  • the present inventors have also devised, in accordance with a second example aspect herein, a computer-implemented method of searching for the presence of a pathology in an image of a portion of an eye acquired by an ocular imaging system.
  • the method comprises receiving image data defining the image, and searching for the presence of at least one of a plurality of different types of pathology in the image by processing the received image data using a learning algorithm trained on image data defining images of healthy eyes, and images of unhealthy eyes each having a respective one of the different types of pathology.
  • the method further comprises, in case at least one of the plurality of different types of pathology is found to be present in the image, processes of: selecting, for each of at least one type of pathology found to be present in the image, a respective one of a plurality of different types of measurement modality which is to be used to perform a measurement on the portion of the eye; generating, for each of the at least one type of pathology found to be present in the image, a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the portion of the eye; and receiving measurement data of the measurement performed by the eye measurement apparatus of each selected measurement modality.
  • the present inventors have further devised, in accordance with a third example aspect herein, a computer program which, when executed by a computer, causes the computer to perform the method according to the first or the second example aspects herein.
  • the present inventors have further devised, in accordance with a fourth example aspect herein, a non-transitory computer-readable storage medium storing the computer program according to the third example aspect herein.
  • the present inventors have further devised, in accordance with a fifth example aspect herein, a signal carrying the computer program according to the third example aspect herein.
  • the present inventors have further devised, in accordance with a sixth example aspect herein, an apparatus for searching for a region indicative of a pathology in an image of a portion of an eye acquired by an ocular imaging system.
  • the apparatus comprises a receiver module configured to receive image data defining the image.
  • the apparatus further comprises a search module configured to search for the region in the image by processing the received image data using a learning algorithm trained on image data defining images of the portion of healthy eyes, and image data defining images of the portion of unhealthy eyes each having at least one region that is indicative of the pathology.
  • the apparatus further comprises an instruction generating module configured to perform, in response to a region in the image that is indicative of the pathology being found by the search module, processes of: determining a location of the region in the image; and generating an instruction for an eye measurement apparatus to perform a measurement on the portion of the eye to generate measurement data, using a reference point based on the determined location for setting a location of the measurement on the portion of the eye.
  • the receiver module is further configured to receive the measurement data from the eye measurement apparatus.
  • the present inventors have further devised, in accordance with a seventh example aspect herein, an apparatus for searching for the presence of a pathology in an image of a portion of an eye acquired by an ocular imaging system.
  • the apparatus comprises a receiver module configured to receive image data defining the image and a search module configured to search for the presence of at least one of a plurality of different types of pathology in the image by processing the received image data using a learning algorithm trained on image data defining images of healthy eyes, and images of unhealthy eyes each having a respective one of the different types of pathology.
  • the apparatus further comprises an instruction generating module configured to perform, in response to at least one of the plurality of different types of pathology being found to be present in the image by the search module, processes of: selecting, for each of at least one type of pathology found to be present in the image, a respective one of a plurality of different types of measurement modality which is to be used to perform a measurement on the portion of the eye; and generating, for each of the at least one type of pathology found to be present in the image, a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the portion of the eye.
  • the receiver module is further configured to receive measurement data of the measurement performed by the eye measurement apparatus of each selected measurement modality.
  • FIG. 1 is a schematic illustration of an apparatus for searching for a region indicative of a pathology in an ocular image, in accordance with a first example embodiment herein.
  • FIG. 2 is a block diagram illustrating an example signal processing hardware configuration of the apparatus of FIG. 1 , according to an example embodiment herein.
  • FIG. 3 is a flow diagram illustrating a process by which the apparatus of FIG. 1 searches for a region indicative of a pathology in an ocular image, in accordance with the first example embodiment herein.
  • FIG. 4( a ) is a schematic illustration of an image of a portion of the eye acquired by an ocular imaging system.
  • FIG. 4( b ) is an annotated version of the image shown in FIG. 4( a ) , showing a region which is indicative of a pathology that has been found, in accordance with an example aspect herein.
  • FIG. 5 is a schematic illustrating how a learning algorithm may be trained on image data defining images of the portion of healthy eyes, and image data defining images of the portion of unhealthy eyes each having at least one region that is indicative of a pathology, in accordance with a first example aspect herein.
  • FIG. 6 is a schematic illustrating how a learning algorithm may be trained on image data defining images of a portion of healthy eyes, and image data defining images of the portion of unhealthy eyes each having at least one region that is indicative of a respective one of a plurality of different types of pathology, in accordance with a second example aspect herein.
  • FIG. 7 is a schematic illustration of a convolutional neural network comprising artificial neurons in an input layer, a hidden layer, and an output layer.
  • FIG. 8 is a flow diagram illustrating a process by which a search module of the apparatus of FIG. 1 may search for a region indicative of pathology in the image, according to an example embodiment herein.
  • FIG. 9 is a schematic illustration of an apparatus for searching for the presence of a pathology in an ocular image, in accordance with an example embodiment herein.
  • FIG. 10 is a flow diagram illustrating a process by which the apparatus of FIG. 9 searches for the presence of a pathology in an ocular image, in accordance with an example embodiment herein.
  • an ocular image is processed to automatically detect a pathology in the ocular image using the image processing techniques hereinafter described, and then a further measurement on the eye is automatically instructed in order to acquire supplementary data that can allow the presence of the pathology to be determined (and/or to allow the pathology to be characterised) with a higher degree of confidence.
  • the eye may be imaged using a different imaging modality (for example, optical coherence tomography, OCT), or its functional response to light stimulation measured.
  • an OCT scan of the region may be instructed, and thickness or volume measurements may be obtained from the data of the OCT scan in order to confirm the presence of the macular oedema and, where present, optionally diagnose its severity.
  • Embodiments can thus facilitate fast and reliable detection of pathologies for early detection of disease, their advantages being most pronounced where the ocular image is a wide-field or ultra-wide-field image covering a large part of the retina or other portion of the eye.
  • FIG. 1 is a schematic illustration of an apparatus 100 according to a first example embodiment, for searching for a region indicative of a pathology in an image of a portion of an eye acquired by an ocular imaging system (shown at 520 in the schematic illustration of FIG. 5 ).
  • the apparatus 100 comprises a receiver module 110 configured to receive image data defining an image produced by a ocular imaging system 520 , and a search module 120 configured to automatically search for at least one region indicative of a pathology in the image, by processing the received image data using a learning algorithm which is described in detail below.
  • the apparatus 100 further comprises an instruction generating module 130 configured to automatically perform, in response to a region in the image that is indicative of the pathology being found by the search module 120 , processes of: determining a location of the region in the image and generating an instruction for an eye measurement apparatus to perform a measurement on the portion of the eye to generate measurement data, using a reference point based on the determined location for setting a location of the measurement on the portion of the eye. In cases where two or more such regions are found by the search module 120 , these processes may be performed for each of the regions.
  • the receiver module 110 is further configured to receive the measurement data from the eye measurement apparatus.
  • the apparatus 100 may, as in the present illustrated embodiment, comprise a display control signal generator 140 (this optional component being shown by the dashed lines in FIG. 1 ).
  • a region indicative of a pathology may be a region of the image defined by the received image data which contains image features indicative of a lesion, structural damage or any other abnormality that would not be present in a similarly-acquired image of a healthy eye.
  • the apparatus 100 may be configured to search for a region indicative of any pathology that may occur in the eye, which may include (but is not limited to), for example, glaucoma, moderate diabetic retinopathy, severe diabetic retinopathy, tumour, drusen, oedema and atrophy.
  • the receiver module 110 is configured to receive image data defining an image of a portion of a retina of the eye, which has been generated by the ocular imaging system 520 .
  • the received image data may define an image of a portion of the eye other than the retina, for example a portion of the anterior segment of the eye, or a portion of the posterior segment of the eye.
  • the received image data may, as in the present example embodiment, define a two-dimensional image, or it may alternatively define a three-dimensional image of the imaged portion of the eye.
  • the received image data may be provided in any suitable format (whether compressed or uncompressed) known to those skilled in the art.
  • the image data received by the receiver module 110 is of a first imaging modality (discussed in more detail below), and represents a result of imaging the retina of the eye using appropriately selected values for imaging parameters, which may include imaging resolution, aperture size, and wavelength.
  • the ocular imaging system 520 may be any ocular imaging system that is suitable for imaging the retina (or other selected portion) of the eye.
  • the ocular imaging system 520 may, for example, be a fundus camera, or a type of scanning imaging system.
  • the ocular imaging system 520 of the present example embodiment is a scanning imaging system in the exemplary form of a scanning laser ophthalmoscope (SLO), which is configured to acquire images of the retina of a subject's eye.
  • SLO of the present example embodiment is configured to capture autofluorescence (AF) images (it may be configured to capture Red-Green (RG) reflectance images or images from other fluorescence modes), although it may alternatively or additionally be configured to acquire one or more other types of images.
  • AF autofluorescence
  • RG Red-Green
  • the SLO may, for example, be an ultra-wide field SLO (UWF-SLO) capable of generating an ultra-wide field image of up to 80% of a retinal surface.
  • the ocular imaging system 520 may be of another imaging modality, for example an optical coherence tomography (OCT) scanner, in which case the image processing techniques described herein are applicable to the tomographic images acquired by the OCT scanner.
  • OCT optical coherence tomography
  • the ocular imaging system 520 may be a combined SLO-OCT scanner, in which case the image processing techniques described herein are applicable to both the SLO retinal scans and the OCT scans acquired by the combined SLO-OCT scanner.
  • the receiver module 110 is configured to receive image data defining the image acquired by the ocular imaging system 520 by any suitable means known to those versed in the art.
  • the receiver module 110 may receive the image data from the ocular imaging system 520 via a direct communication link (which may be provided by any suitable wired or wireless connection, e.g. a Universal Serial Bus (USB) or a BluetoothTM connection), or an indirect communication link (which may be provided by a network comprising a Local Area Network (LAN), a Wide Area Network (WAN) and/or the Internet).
  • the image data may be received by the receiver module 110 receiving (e.g. by reading from a storage medium such as a CD or hard disk, or receiving via a network such as the Internet) such image data after it has been acquired by the ocular imaging system.
  • the image data may be received by the receiver module 110 (and may furthermore subsequently be processed to search for a region indicative of a pathology in an image of a portion of an eye, as described below) as this image data is being generated by the ocular imaging system, i.e. the image data may be acquired “on the fly”, without waiting for the ocular imaging system 520 to finish generating all of the image data that forms the image of the portion of the retina.
  • the receiver module 110 is configured to receive all of the image data defining the image of the portion of the eye before the search module 120 begins to process this image data.
  • the display control signal generator 140 may be arranged to generate display control signals for controlling a display device (as shown at 215 in FIG. 2 ), such as an LCD screen or other type of visual display unit, to display both the location of the region indicative of the pathology in the image of the portion of the eye, and a representation of the received measurement data.
  • a display device as shown at 215 in FIG. 2
  • the display control signal generator 140 may be arranged to generate display control signals for controlling a display device (as shown at 215 in FIG. 2 ), such as an LCD screen or other type of visual display unit, to display both the location of the region indicative of the pathology in the image of the portion of the eye, and a representation of the received measurement data.
  • FIG. 2 is a schematic illustration of a programmable signal processing hardware 200 , which may, as in the present example embodiment, be configured to function as the apparatus 100 of FIG. 1 .
  • the programmable signal processing hardware 200 comprises a communication interface (I/F) 210 for receiving the image data described above, generating the instruction for the eye measurement apparatus 300 to perform the measurement on the portion of the eye to generate the measurement data, receiving the measurement data from the eye measurement apparatus 300 , and, optionally, for outputting display control signals for controlling the display device 215 to display both the image of the portion of the eye and a representation of the measurement data.
  • the signal processing apparatus 200 further comprises a processor (e.g.
  • the instruction store 240 may comprise a ROM (e.g. in the form of an electrically-erasable programmable read-only memory (EEPROM) or flash memory) which is pre-loaded with the computer-readable instructions.
  • EEPROM electrically-erasable programmable read-only memory
  • the instruction store 240 may comprise a RAM or similar type of memory, and the computer-readable instructions of the computer program can be input thereto from a computer program product, such as a non-transitory, computer-readable storage medium 250 in the form of a CD-ROM, DVD-ROM, etc. or a computer-readable signal 260 carrying the computer-readable instructions.
  • the computer program when executed by the processor, causes the processor to execute at least one of the methods of searching for a region indicative of a pathology, or the presence of a pathology, in an image of a portion of an eye acquired by an ocular imaging system described herein.
  • the apparatus 100 may alternatively be implemented in non-programmable hardware, such as an application-specific integrated circuit (ASIC).
  • ASIC application-specific integrated circuit
  • a combination 270 of the hardware components shown in FIG. 2 comprising the processor 220 , the working memory 230 and the instruction store 240 , is configured to perform functions of the search module 120 and instruction generating module 130 , which functions will now be described in detail below.
  • the apparatus 100 comprises a display control signal generator 140
  • the functionality of this optional component also be provided by the combination 270 of the hardware components, together with the communication I/F 210 .
  • the apparatus 100 automatically processes image data defining an image of the portion of an eye acquired by an ocular imaging system 520 to find, and record the location in the image of, each region that is indicative of pathology, and generates a corresponding instruction for an eye measurement apparatus 300 to perform a measurement on the portion of the eye to generate measurement data, using a reference point based on the recorded location of the respective region for setting a respective location of the measurement on the portion of the eye.
  • the apparatus 100 thus acquires both the location of each image region suspected to contain a pathology, and additional measurement data related to the respective region.
  • FIG. 3 is a flow diagram illustrating a process by which the apparatus 100 of FIG. 1 searches for a region that is indicative of a pathology in an image of a portion of an eye acquired by an ocular imaging system 520 , in accordance with the first example embodiment herein.
  • the receiver module 110 receives image data of a first imaging modality, which has been acquired by the ocular imaging system 520 .
  • the imaging modality of the ocular imaging system may, for example, take one of the many different forms known to those versed in the art, including OCT, colour fundus photography, fluorescein angiography (FA), indocyanine green angiography (ICG) and autofluoresence (AF), among others.
  • the received image data defines the image of a portion of the eye (in this example embodiment, the retina) acquired by the ocular imaging system 520 .
  • FIG. 4( a ) is a schematic illustration of an image 400 of the portion of the retina defined by the received image data that has been acquired by an ocular imaging system 520 .
  • the search module 120 searches for the region ( 410 in FIG. 4( a ) ) indicative of a single type of pathology in the image 400 by processing the received image data using a learning algorithm which has been trained on image data defining images of the retina of healthy eyes, and image data defining images of the retina of unhealthy eyes each having at least one region that is indicative of the pathology.
  • the location in the image 400 of each region 410 found in the search is recorded by the search module 120 , for example in the working memory 230 in case the apparatus 100 is implemented in the programmable signal processing hardware 200 of FIG. 2 .
  • FIG. 5 is a schematic representation illustrating how a learning algorithm may be trained on image data 502 defining images of healthy eyes, and image data 501 defining images of unhealthy eyes.
  • the learning algorithm 530 is configured to learn from, and make predictions based on, input data by building a model 540 from an example training set 500 of input data, comprising the image data 502 defining images of the retina of healthy eyes, and the image data 501 defining images of the retina of unhealthy eyes.
  • image data 501 defines images of the portion of unhealthy eyes, each of which has a region indicative of severe diabetic retinopathy.
  • the images defined by the image data 501 in the example training set 500 may be collected by acquiring images of the retinas of multiple subjects.
  • each image is of the same portion of the eye (or of substantially the same portion of the eye 510 or of a part of the eye containing the same portion) as the image defined by the received image data.
  • each image defined by the image data in the example training set 500 is acquired by the ocular imaging system 520 or by the same type of ocular imaging system and operating in the same imaging modality.
  • image data 501 defines images of the portion of unhealthy eyes, each of which has a region indicative of severe diabetic retinopathy.
  • each example image in the example training set 500 is a pair consisting of input image data defining an image of the portion of the eye and a desired output value indicating whether the image is of a portion of a “healthy” or “unhealthy” eye.
  • the supervised learning algorithm 530 analyses the image data in the example training set 500 and produces a model 540 , which can be used to classify new unseen image data defining an image of the portion of the eye as “healthy” or “unhealthy”.
  • the model 540 cannot distinguish between pathologies. It can only determine whether a region indicative of the pathology, for example, severe diabetic retinopathy, is present or not.
  • the search module 120 may, in a modification of the present example embodiment, search for the region 410 in the image by searching for a region 410 in the image that is indicative not of a single type of pathology, but of one of a plurality of different types of pathology, by processing the received image data using the learning algorithm. Therefore, in the modification of the present example embodiment, the learning algorithm may be trained on image data defining images of the portion of unhealthy eyes each having a respective one of the different types of pathology.
  • FIG. 6 is a schematic illustrating how the modified learning algorithm 630 may be trained on image data 604 defining images of the retina of healthy eyes, and image data 601 , 602 , 603 defining images of the retina of unhealthy eyes, each having at least one region that is indicative of a respective one of a plurality of different types of pathology.
  • image data 601 defines images of the portion of unhealthy eyes which have glaucoma
  • image data 602 defines images of the portion of unhealthy eyes which have moderate diabetic retinopathy
  • image data 603 defines images of the portion of unhealthy eyes which have severe diabetic retinopathy.
  • the learning algorithm 630 is configured to learn from, and make predictions based on, input data by building a model 640 from an example training set 600 of input data, comprising the image data 604 defining images of the retina of healthy eyes, and the image data 601 , 602 , 603 defining images of the retina of unhealthy eyes.
  • the images defined by the image data 601 to 604 in the example training set 600 may be collected by acquiring images of the retinas of multiple subjects. More generally, each image is of the same portion of the eye (or of substantially the same portion of the eye 510 or of a part of the eye containing the same portion) as the image defined by the received image data.
  • each image defined by the image data 610 to 604 in the example training set 600 is acquired by the ocular imaging system 520 or by the same type of ocular imaging system and operating in the same imaging modality.
  • the learning algorithm 630 may be a supervised learning algorithm.
  • each example image of the portion of the eye 510 in the example training set 600 is associated with an indicator indicating whether that image is of a portion of a “healthy” or an “unhealthy” eye and, in cases where the image is of a portion of a “unhealthy” eye, also a second indicator indicating which one of the plurality of pathologies is present in the image.
  • the supervised learning algorithm analyses the image data in the example training set 600 and produces a model 640 , which can be used to classify new (previously unseen) image data defining an image of the portion of the eye as one of, for example: “healthy”; “unhealthy—glaucoma”; “unhealthy—moderate diabetic retinopathy”; and “unhealthy—severe diabetic retinopathy”.
  • the apparatus 100 may be adapted to classify additional pathologies by expanding the training set 600 to include, for each of the additional pathologies, image data defining images of the retina (or other portion) of unhealthy eyes having that pathology, and associated indicators as described above.
  • the training set 600 may be expanded to include image data defining images of the portion of unhealthy eyes having a tumour and/or image data defining images of the portion of unhealthy eyes having oedema.
  • any of the image data 601 , 602 , 603 defining images of the portion of unhealthy eyes may be removed or replaced with image data defining images of the portion of unhealthy eyes having a different pathology.
  • a revised version of the model 640 may then be produced based on the modified training set 600 .
  • the supervised learning algorithm 630 may, as in the present example embodiment, be a neural network. Neural networks automatically generate identifying characteristics by processing the input data, such as the image data in the example training set 600 , without any prior knowledge.
  • the neural network may, as in the present example embodiment, be a convolutional neural network.
  • Convolutional neural networks are particularly suitable to image and video recognition tasks.
  • a convolution neural network consists of an input layer and an output layer, as well as multiple hidden layers.
  • Each of the layers is composed of a plurality of artificial neurons (labelled A to F in FIG. 7 ), and each layer may perform different kinds of transformations on their inputs.
  • Each artificial neuron may be connected to multiple artificial neurons in adjacent layers.
  • the output of each artificial neuron is computed by some non-linear function of the sum of its inputs.
  • Artificial neurons and the connections therebetween typically have respective weights (WAD, WAE, etc. in FIG. 7 ) which determined the strength of the signal at a given connection. These weights are adjusted as learning proceeds, thereby adjusting the output of the convolutional neural network. Signals travel from the first layer (the input layer), to the last layer (the output layer), and may traverse the layers multiple times.
  • the output of the neural network may be viewed as a probability of the input image data containing identifying characteristics of the pathology and the classification may, as in the present example embodiment, comprise determining whether the output of the trained model 640 exceeds a predetermined threshold.
  • the predetermined threshold may represent an acceptably low probability that the input image data contains identifying characteristics of a pathology and, therefore, a high probability that the eye of the subject is healthy.
  • the search module 120 may be configured, as part of process S 12 of FIG. 3 , to search for the region 410 indicative of the pathology for in the image by deconstructing the neural network.
  • the trained model 640 classifies this image as “healthy”, or as “unhealthy” and having a particular pathology (that is, the model 640 determines whether the input image data contains identifying characteristics of a pathology or not).
  • Conventional convolutional neural networks do not output an explanation of this classification. Accordingly, the convolutional neural network can be deconstructed, that is, processed in order to determine which input variables of the input image data (that is, pixels of the image 400 ) were relevant to the output of the neural network.
  • the input variables most relevant to the classification as “unhealthy” and having a certain pathology of the image 400 defined by the received image data correspond to the region 410 (or regions) that are indicative of the pathology.
  • the search module 120 may, as in the present example embodiment, be configured not to deconstruct the neural network.
  • the search module 120 may, as in the present example embodiment, be configured to deconstruct the convolutional neural network by performing the processes illustrated in the flow diagram of FIG. 8 .
  • the search module 120 performs, for each of a plurality of different sections of the image 400 , a process of masking the section of the image to generate a masked image.
  • the search module 120 performs, for each of a plurality of different sections of the image 400 , a process of searching for the region in the masked image by processing image data defining the masked image using the learning algorithm.
  • the search module 120 performs, for each of a plurality of different sections of the image 400 , a process of determining a difference between a first result which is a result of the search performed using the image data defining the masked image, and a second result which is a result of a search performed using the received image data (that is, the image data as received, in which the section of the image is not masked).
  • the search module 120 determines, as the location of the region 410 in the image 400 , a location of a section for which the determined difference is largest.
  • the search module 120 may be configured to deconstruct the convolutional neural network by:
  • Taylor decomposition is a method of explaining individual neural network predictions by decomposing the output of the neural network on its input variables.
  • a function ⁇ (x) quantifies the presence of a certain type of object, here a region 410 indicative of a pathology, in the image 400 .
  • a function value ⁇ (x)>0 expresses its presence with a certain degree of certainty, or in a certain amount.
  • the Taylor decomposition method assigns each pixel p in the image a relevance score R p (x), that indicates for an image x to what extent the pixel p contributes to explaining the classification decision ⁇ (x).
  • the search module 120 may be configured to deconstruct the convolutional neural network by determining a deconvolution of the convolutional neural network.
  • the convolution neural network may be deconstructed by training a multi-layer deconvolution network which takes, as an input, the output of the convolutional neural network and provides, as an output, a probability map of the same size as the image input to the convolutional neural network which indicates the probability of each pixel belonging to one of the predefined classes.
  • the location of the region may be determined as the location of the pixels having the highest probability of belonging to one of the predetermined classes (that is, the highest probability of having one of the plurality of different types of pathology).
  • the search module 120 may, as in the present example embodiment, be configured to search for the region 410 indicative of a pathology in the image by identifying, as the region 410 indicative of a pathology, the pixels of the image 400 defined by the received image data that are most relevant to the output of the neural network.
  • the search module 120 determines whether a region in the image that is indicative of the pathology has been found as a result of the search performed in process S 12 . If a region in the image that is indicative of the pathology has been found as a result of the search performed in process S 12 , the process proceeds to S 14 , otherwise the process ends.
  • the instruction generating module 130 determines the location in the image of the region 410 which has been found by the search module 120 .
  • the location may, for example, be recorded by the search module 120 in the working memory 230 of the hardware configuration 200 illustrated in FIG. 2 , and read from the working memory by the instruction generating module 130 in S 14 .
  • the determined location may, as illustrated in FIG. 4( a ) , serve to center or otherwise position an overlay graphical element, such as the boundary box 440 illustrated in FIG. 4( b ) , over the region 410 of the image 400 found to contain the pathology.
  • the instruction generating module 130 generates an instruction for an eye measurement apparatus 300 to perform a measurement on the portion of the eye to generate measurement data, using a reference point based on the determined location for setting a location of the measurement on the portion of the eye.
  • the instruction generated in process S 16 of FIG. 3 may be provided to the eye measurement apparatus 300 by any suitable means known to those versed in the art.
  • the determined location is a location in the image 400 of a group of one or more pixels constituting the region 410 .
  • the location recorded by the search module 120 in S 12 can be transformed into a corresponding set of one or more control parameters for steering the eye measurement apparatus 300 to perform its measurement at substantially the same location in/on the eye as that imaged by the ocular imaging system 520 . This can be done in one of a number of different ways.
  • the instruction generating module 130 may use a mapping between pixel coordinate values from images acquired by the SLO and corresponding values of the control parameters, which may be provided in the form of a look-up table or a function defined by a set of parameters, for example.
  • the mapping may be determined by calibration, using techniques known to those skilled in the art.
  • the receiver module 110 receives the measurement data from the eye measurement apparatus 300 , and the process then ends.
  • the eye measurement apparatus 300 may be of any kind of apparatus that allows a supplementary measurement of the subject's eye to be made, which could be used to verify that the region found by the search module 120 does (or is highly likely to) contain an imaged pathology.
  • the eye measurement apparatus 300 may, for example, measure a functional response of the subject's eye to stimulation by light in response to an instruction generated by the instruction generating module 130 , and communicate data representing the measured functional response to the receiver module 110 .
  • the eye measurement apparatus 300 may, as in the present embodiment, be configured to perform, as the measurement on the retina (or other selected portion of the eye, as noted above), an image capture process of a second imaging modality to image a region of the retina, using the aforementioned reference point for setting, as the location of the measurement, a location of the region of the retina to be imaged in the image capture process, wherein the second imaging modality is different to the first imaging modality (i.e. SLO in the present example embodiment).
  • the instruction generating module 120 generates in S 16 of FIG.
  • an instruction for the eye measurement apparatus 300 to perform the measurement on the portion of the eye an instruction for the eye measurement apparatus 300 to perform the image capture process using the reference point for setting, as the location of the measurement, a location of a region in the portion of the eye to be imaged in the image capture process.
  • the receiver module 110 then receives in process S 18 of FIG. 3 , as the measurement data, image data of the second imaging modality acquired by the eye measurement apparatus 300 .
  • a single, multi-modal ocular imaging system capable of acquiring ocular images of two or more different imaging modalities (e.g. a combined SLO-OCT imaging system) may provide the functionality of the ocular imaging system 520 and the eye measurement apparatus 300 .
  • the ocular imaging system 520 and the eye measurement apparatus may alternatively be provided as separate components.
  • the mapping discussed above may rely on the subject's eye being positioned and oriented in relation to the eye measuring apparatus in a predetermined way. While this may be a good assumption in any practical applications, it may not hold true in all cases. Where this assumption does not hold, it may be preferable to define the location of the region found by the search module 120 in relation to an anatomical feature (e.g. the fovea) that is recognisable in both the image acquired by the ocular imaging system 520 and the image of a different modality that is acquired by the eye measurement apparatus 300 of the present example embodiment. The location of the region relative to the position of the fovea in the image 400 may then be transformed into a corresponding location relative to the position of the fovea in an image acquired by the eye measurement apparatus 300 .
  • anatomical feature e.g. the fovea
  • the reference point may be used to set the location of the measurement on the portion of the eye in any suitable way.
  • the eye measurement apparatus 300 may be controlled to image (or more generally perform any other kind of measurement on) a region of a predetermined size that is centred on a location on the subject's retina corresponding to the reference point.
  • the instruction generating module 130 performs, in response to a region 410 in the image 400 that is indicative of one of the plurality of different types of pathology being found by the search module 120 , a process of selecting, for the one of the plurality of different types of pathology and as the second imaging modality, a respective one of a plurality of different types of imaging modality which is to be used to perform the image capture process on the retina of the eye.
  • the instruction generating module 130 may, for example, select the respective one of a plurality of different types of imaging modality, which is to be used to perform the image capture process on the subject's retina, by consulting a look-up table (LUT) in which an indicator of each of the plurality of different types of pathology is stored in association with a respective indicator of one of the plurality of different types of imaging modality, and using the one of the plurality of different types of pathology found by the search module 120 as a search key.
  • LUT look-up table
  • the instruction generating module 130 may generate an instruction for the eye measurement apparatus 300 to perform, as the image capture process of the second imaging modality, an OCT scan of the region on the retina of the subject's eye.
  • the receiver module 110 may then receive in S 18 of FIG. 3 , as the measurement data, image data of the OCT scan, which the apparatus 100 can process to obtain complementary data that is useful for detecting the glaucoma (and optionally estimating its severity), namely a measurement of a retinal nerve fibre layer and/or an optic nerve head of the eye.
  • the instruction generating module 130 may generate an instruction for the eye measurement apparatus 300 to perform, as the image capture process of the second imaging modality, an OCT scan of a region of a retina of the eye.
  • the receiver module 110 may then receive in S 18 of FIG. 3 , as the measurement data, image data of the OCT scan, which the apparatus 100 can process to obtain complementary data that is useful for detecting the severe diabetic retinopathy (and optionally estimating its severity), namely measurements of macular thickness.
  • the instruction generating module 130 may generate an instruction for the eye measurement apparatus 300 to perform, as the image capture process of the second imaging modality, a high-density OCT B-scan of the region in the retina of the eye.
  • the receiver module 110 may then receive in S 18 of FIG. 3 , as the measurement data, image data of the high-density OCT B-scan, which can be useful for detecting the tumour (and optionally estimating its size).
  • the instruction generating module 130 may generate an instruction for the eye measurement apparatus 300 to perform, as the image capture process of the second imaging modality, an OCT B-scan of the region in the retina of the eye.
  • the receiver module 110 may then receive in S 18 of FIG. 3 , as the measurement data, image data of the OCT B-scan, which can be useful for detecting the presence of drusen or estimating their size and/or number. This data can be valuable in diagnosing age-related macular degeneration at an early stage, for example.
  • the instruction generating module 130 may generate an instruction for the eye measurement apparatus 300 to perform, as the image capture process of the second imaging modality, an OCT scan of the region in the retina of the eye.
  • the receiver module 110 may then receive in S 18 of FIG. 3 , as the measurement data, image data of the OCT scan, which can be useful for detecting the oedema (or atrophy, as the case may be), and optionally estimating its severity.
  • the search module 120 searches for a region 410 in the image that is indicative of one of a plurality of different types of pathology by processing the received image data using the learning algorithm, and the instruction generating module 130 performs, in response to a region 410 in the image 400 that is indicative of one of the plurality of different types of pathology being found by the search module 120 , a process of selecting, for the one of the plurality of different types of pathology and as the second imaging modality, a respective one of a plurality of different types of imaging modality which is to be used to perform the image capture process on the retina of the eye.
  • the eye measurement apparatus 300 need not be configured to acquire an image of the retina (or, more generally, any other selected portion of the eye) but may instead perform a different kind of measurement on the eye.
  • the eye measurement apparatus 300 may be multi-modal in the sense of being operable in a selected one of a plurality of different measurement modalities.
  • the instruction generating module 130 may perform, in response to a region 410 in the image 400 that is indicative of one of the plurality of different types of pathology being found by the search module 120 , processes of: selecting, for the one of the plurality of different types of pathology, a respective one of a plurality of different types of measurement modality for a measurement to be performed on the eye; and generating, as the instruction for the eye measurement apparatus 300 to perform the measurement on the portion of the eye, an instruction for an eye measurement apparatus 300 of the selected measurement modality to perform the measurement on the portion of the eye, using the reference point for setting the location of the measurement on the portion of the eye.
  • the eye measurement apparatus 300 may be an ocular imaging apparatus of the same imaging modality as the ocular imaging system 520 which, similar to the first embodiment, does not distinguish between different types of pathology.
  • the receiver module 110 is configured to receive image data representing a result of imaging the portion of the eye using a first value of an imaging parameter, the imaging parameter being an imaging resolution, an aperture size or wavelength used in the imaging.
  • the instruction generating module 130 is configured to perform, in response to a region 410 in the image 400 that is indicative of the pathology being found by the search module 120 , a process of generating, as the instruction for the eye measurement apparatus 300 to perform the measurement on the portion of the eye, an instruction for the eye measurement apparatus 300 to perform an image capture process using a second value of the imaging parameter to image a region in the portion of the eye, again using the reference point for setting, as the location of the measurement, a location of a region in the portion of the eye to be imaged in the image capture process, wherein the second value of the imaging parameter is different from the first value of the imaging parameter.
  • the receiver module 110 is configured to receive from the eye measurement apparatus 300 , as the measurement data, the image data representing the result of imaging the region in the portion of the eye using the second value of the imaging parameter and the received image data and the received measurement data are of the same imaging modality.
  • the ocular imaging system 520 may be configured to acquire image data defining the image by imaging the portion of the eye at a first imaging resolution
  • the instruction generating module 130 may be configured to generate, in response to a region in the image that is indicative of pathology being found by the search module 120 , an instruction for the eye measurement apparatus 300 to perform an image capture process at a second, higher imaging resolution to image the region in more detail.
  • FIG. 9 is a schematic illustration of an apparatus 800 for searching for the presence of a pathology in an image of a portion of an eye acquired by an ocular imaging system ( 520 in FIG. 5 ), in accordance with a second example embodiment herein.
  • the apparatus 800 of the second example embodiment differs from the apparatus 100 of FIG. 1 in that the apparatus of FIG. 9 is not required to determine the location of a region indicative of a pathology in the received image 400 .
  • the apparatus 800 comprises a receiver module 810 configured to receive image data defining an image produced by a ocular imaging system 520 , and a search module 820 configured to search for the presence of at least one of a plurality of different types of pathology in the image.
  • the apparatus 800 further comprises an instruction generating module 830 configured to perform, in response to at least one of the plurality of different types of pathology being found to be present in the image by the search module 820 , processes of: selecting, for each of at least one type of pathology found to be present in the image, a respective one of a plurality of different types of measurement modality which is to be used to perform a measurement on the portion of the eye; and generating, for each of the at least one type of pathology found to be present in the image, a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the portion of the eye.
  • the receiver module 810 is further configured to receive measurement data of the measurement performed by the eye measurement apparatus of each selected measurement modality.
  • the apparatus 800 may, as in the present illustrated embodiment comprise a display control signal generator 840 (this optional component being shown by the dashed lines in FIG. 9 ).
  • the apparatus 800 may be configured to search for the presence of any pathology, (including, for example, glaucoma, moderate diabetic retinopathy, severe diabetic retinopathy, tumour, drusen, oedema and atrophy), as discussed above in relation to the first embodiment.
  • pathology including, for example, glaucoma, moderate diabetic retinopathy, severe diabetic retinopathy, tumour, drusen, oedema and atrophy
  • the receiver module 810 is configured to receive image data defining an image of a portion of a retina of the eye, which has been generated by the ocular imaging system 520 .
  • the received image data may define an image of a portion of the eye other than the retina, for example a portion of the anterior segment of the eye, or a portion of the posterior segment of the eye.
  • the ocular imaging system 520 may, as in the present example embodiment, be a scanning laser ophthalmoscope. Alternatively, the ocular imaging system may be any ocular imaging system described above in relation to the first embodiment.
  • the receiver module 810 may be configured to receive image data defining the image acquired by the ocular imaging system 520 by any of the means discussed above in relation to the first embodiment.
  • the apparatus 800 automatically processes image data defining an image of the portion of an eye acquired by an ocular imaging system 520 to search for the presence of at least one of a plurality of different types of pathology in the image and, when at least one of the plurality of different types of pathology is found to be present in the image, selects for each of at least one type of pathology found to be present in the image a respective one of a plurality of different types of measurement modality which is to be used to perform a measurement on the portion of the eye.
  • a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the portion of the eye is then generated for each of the at least one type of pathology found to be present in the image.
  • the apparatus 800 thus acquires both the image suspected to contain a particular type of pathology, and additional measurement data of a measurement modality that is related to the suspected type of pathology.
  • These different but complementary kinds of information that are automatically acquired by the apparatus 800 may allow any pathology-containing ocular image to be identified quickly and with a high degree of confidence.
  • the apparatus 800 of FIG. 9 may be implemented by a signal processing hardware configuration, such as that shown in the FIG. 2 , or by any other suitable means.
  • FIG. 10 is a flow diagram illustrating a process by which the apparatus 800 of FIG. 9 searches for the presence of a pathology in an image of a portion of an eye acquired by an ocular imaging system 520 , in accordance with the second example embodiment herein.
  • the receiver module 810 receives image data defining the image of the portion of the eye acquired by the SLO (as an example of the ocular imaging system 520 ).
  • the search module 820 searches for the presence of at least one of a plurality of different types of pathology in the image by processing the received image data using a learning algorithm trained on image data defining images of healthy eyes, and images of unhealthy eyes each having a respective one of the different types of pathology.
  • the learning algorithm may be trained as discussed above in relation to FIG. 6 in order to classify input image data defining an image of the portion of an eye as “healthy” or as “unhealthy” and having a particular pathology.
  • the apparatus 800 is not required to record a location of a region indicative of pathology in the image. Therefore, the apparatus 800 may, as in the present example embodiment, not process the learning algorithm in order to determine which input variables of the input image data (that is, pixels of the image defined by the received image data) were relevant to the output (that is, the finding that one of a plurality of different types of pathology is present). Alternatively, in other embodiments, such processing may optionally be carried out by the apparatus 800 as part of S 22 of FIG. 10 in order to identify a reference point.
  • the search module 820 determines whether at least one of the different types of pathology has been found to be present in the image, as a result of the search performed in process S 22 . If at least one of the different types of pathology has been found to be present in the image as a result of the search performed in process S 22 , the process proceeds to S 24 , otherwise the process ends.
  • the instruction generating module 830 performs the process of selecting, for each of at least one type of pathology found to be present in the image, a respective one of a plurality of different types of measurement modality which is to be used to perform a measurement on the portion of the eye.
  • the instruction generating module 830 may select a respective one of a plurality of different types of measurement modality by any of the means described above in relation to the apparatus 100 of FIG. 1 .
  • the instruction generating module 830 performs the process of generating, for each of the at least one type of pathology found to be present in the image, a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the portion of the eye.
  • the instruction may be generated substantially as discussed above in relation to apparatus 100 of FIG. 1 .
  • the instruction generating module 830 need not use a reference point based on a recorded location of a region indicative of pathology for setting a location of the measurement on the portion of the eye.
  • the instruction generating module 130 may, as in the present example embodiment, generate, for each of the at least one type of pathology found to be present in the image, a respective instruction for an eye measurement apparatus of the respective selected measurement modality to perform the measurement on the same portion of the eye as imaged by the ocular imaging system.
  • the receiver module 810 receives measurement data of the measurement performed by the eye measurement apparatus of each selected measurement modality.
  • the receiver module 810 may receive measurement data by any of the means discussed above in relation of the first embodiment.
  • E6 The computer-implemented method of E1, wherein, in case the region ( 410 ) in the image ( 400 ) that is indicative of the pathology is found in the searching, the method comprises generating (S 16 ), as the instruction for the eye measurement apparatus to perform the measurement on the portion of the eye, an instruction for the eye measurement apparatus to measure a functional response of the eye to light stimulation, using the reference point for setting the location of the measurement which is based on the determined location.
  • E8 The computer-implemented method of any of E1 to E7, further comprising generating instructions for controlling a display unit ( 215 ) to display the location of the region ( 410 ) in the image ( 400 ) of the portion of the eye and a representation of the received measurement data.
  • E10 The computer-implemented method of E9, wherein the supervised learning algorithm comprises a neural network, and the region ( 410 ) indicative of the pathology is searched for in the image ( 400 ) by deconstructing the neural network.
  • E12 The computer-implemented method of E10, wherein the neural network is a convolutional neural network, and the convolutional neural network is deconstructed by:
  • E13 The computer-implemented method of E10, wherein the neural network is a convolutional neural network, and the convolutional neural network is deconstructed by determining a deconvolution of the convolutional neural network.
  • E16 The computer-implemented method of E15, wherein, in case at least one of the plurality of different types of pathology is found to be present in the image ( 400 ),
  • E18 The computer-implemented method of E14, wherein, in case at least one of the plurality of different types of pathology is found to be present in the image ( 400 ), at least one of the instructions generated is an instruction for an eye measurement apparatus ( 300 ) of a selected measurement modality to measure a functional response of the eye to light stimulation.
  • E19 The computer-implemented method of any of E15 to E17, further comprising generating instructions for controlling a display unit ( 215 ) to display the recorded location of the region ( 410 ) in the image ( 400 ) of the portion of the eye and a representation of the received measurement data.
  • E20 The computer-implemented method of any of E15 to E17, wherein the learning algorithm ( 630 ) is a supervised learning algorithm.
  • E21 The computer-implemented method of E20, wherein the supervised learning algorithm comprises a neural network, and a region indicative of one of the different types of pathology found to be present in the image ( 400 ) is searched for in the image by deconstructing the neural network.
  • E24 The computer-implemented method of E21, wherein the neural network is a convolutional neural network, and the convolutional neural network is deconstructed by determining a deconvolution of the convolutional neural network.
  • a computer program which, when executed by a computer, causes the computer to perform a method as set out in at least one of E1 to E24.
  • E33 The apparatus ( 100 ) of E28, wherein, the instruction generating module ( 130 ) is configured to generate, in response to the region ( 410 ) in the image ( 400 ) that is indicative of the pathology being found by the search module ( 120 ), and as the instruction for the eye measurement apparatus ( 300 ) to perform the measurement on the portion of the eye, an instruction for the eye measurement apparatus ( 300 ) to measure a functional response of the eye to light stimulation, using the reference point for setting the location of the measurement which is based on the determined location.
  • E35 The apparatus ( 100 ) of any of E28 to E34, wherein the instruction generating module ( 130 ) is further configured to generate instructions for controlling a display unit ( 215 ) to display the determined location of the region ( 410 ) in the image ( 400 ) of the portion of the eye and a representation of the received measurement data.
  • E36 The apparatus ( 100 ) of any of E28 to E35, wherein the learning algorithm ( 630 ) is a supervised learning algorithm.
  • E37 The apparatus ( 100 ) of E36, wherein the supervised learning algorithm comprises a neural network, and the search module ( 120 ) is configured to search for the region indicative of the pathology in the image by deconstructing the neural network.
  • E38 The apparatus ( 100 ) of E37, wherein the neural network is a convolutional neural network, and the search module ( 120 ) is configured to deconstruct the neural network by:
  • E39 The apparatus ( 100 ) of E37, wherein the neural network is a convolutional neural network, and the search module ( 120 ) is configured to deconstruct the convolutional neural network by:
  • E40 The apparatus ( 100 ) of E37, wherein the neural network is a convolutional neural network, and the search module ( 120 ) is configured to deconstruct the convolutional neural network by determining a deconvolution of the convolutional neural network.
  • the neural network is a convolutional neural network
  • the search module ( 120 ) is configured to deconstruct the convolutional neural network by determining a deconvolution of the convolutional neural network.
  • E42 The apparatus ( 800 ) of E41, wherein the search module ( 820 ) is further configured to perform, in response to finding at least one of the plurality of different types of pathology to be present in the image ( 400 ), a process of:
  • E43 The apparatus ( 800 ) of E42, wherein the instruction generating module ( 830 ) is configured to perform, in response to at least one of the plurality of different types of pathology being found to be present in the image ( 400 ) by the search module ( 820 ), and for each of the at least one of the different types of pathology found to be present in the image ( 400 ), processes of:
  • the receiver module ( 810 ) is configured to receive respective image data of the imaging performed by the eye measurement apparatus ( 300 ) of the selected imaging modality as the measurement data of the measurement performed by the eye measurement apparatus ( 300 ).
  • E45 The apparatus ( 800 ) of E41, wherein the instruction generating module ( 830 ) is configured to generate, in response to at least one of the plurality of different types of pathology being found to be present in the image by the search module ( 820 ), an instruction for an eye measurement apparatus ( 300 ) of a selected measurement modality to measure a functional response of the eye to light stimulation.
  • E46 The apparatus ( 800 ) of any of E42 to E44, wherein the instruction generating module ( 830 ) is further configured to generate instructions for controlling a display unit ( 215 ) to display the recorded location of the region ( 410 ) in the image ( 400 ) of the portion of the eye and a representation of the received measurement data.
  • E48 The apparatus ( 800 ) of E47, wherein the supervised learning algorithm comprises a neural network, and the search module ( 820 ) is configured to search for a region ( 410 ) indicative of one of the different types of pathology found to be present in the image ( 400 ) by deconstructing the neural network.
  • the search module ( 820 ) is configured to search for a region ( 410 ) indicative of one of the different types of pathology found to be present in the image ( 400 ) by deconstructing the neural network.
  • E49 The apparatus of E48, wherein the neural network is a convolutional neural network, and the search module ( 820 ) is configured to deconstruct the neural network by:
  • E50 The apparatus ( 800 ) of E48, wherein the neural network is a convolutional neural network, and the search module ( 820 ) is configured to deconstruct the convolutional neural network by:
  • E51 The apparatus ( 800 ) of E48, wherein the neural network is a convolutional neural network, and the search module ( 820 ) is configured to deconstruct the convolutional neural network by determining a deconvolution of the convolutional neural network.
  • the neural network is a convolutional neural network
  • the search module ( 820 ) is configured to deconstruct the convolutional neural network by determining a deconvolution of the convolutional neural network.
  • Software embodiments of the examples presented herein may be provided as, a computer program, or software, such as one or more programs having instructions or sequences of instructions, included or stored in an article of manufacture such as a machine-accessible or machine-readable medium, an instruction store, or computer-readable storage device, each of which can be non-transitory, in one example embodiment.
  • the program or instructions on the non-transitory machine-accessible medium, machine-readable medium, instruction store, or computer-readable storage device may be used to program a computer system or other electronic device.
  • the machine- or computer-readable medium, instruction store, and storage device may include, but are not limited to, floppy diskettes, optical disks, and magneto-optical disks or other types of media/machine-readable medium/instruction store/storage device suitable for storing or transmitting electronic instructions.
  • the techniques described herein are not limited to any particular software configuration. They may find applicability in any computing or processing environment.
  • computer-readable shall include any medium that is capable of storing, encoding, or transmitting instructions or a sequence of instructions for execution by the machine, computer, or computer processor and that causes the machine/computer/computer processor to perform any one of the methods described herein.
  • Such expressions are merely a shorthand way of stating that the execution of the software by a processing system causes the processor to perform an action to produce a result.
  • Some embodiments may also be implemented by the preparation of application-specific integrated circuits, field-programmable gate arrays, or by interconnecting an appropriate network of conventional component circuits.
  • Some embodiments include a computer program product.
  • the computer program product may be a storage medium or media, instruction store(s), or storage device(s), having instructions stored thereon or therein which can be used to control, or cause, a computer or computer processor to perform any of the procedures of the example embodiments described herein.
  • the storage medium/instruction store/storage device may include, by example and without limitation, an optical disc, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory, a flash card, a magnetic card, an optical card, nanosystems, a molecular memory integrated circuit, a RAID, remote data storage/archive/warehousing, and/or any other type of device suitable for storing instructions and/or data.
  • some implementations include software for controlling both the hardware of the system and for enabling the system or microprocessor to interact with a human user or other mechanism utilizing the results of the example embodiments described herein.
  • Such software may include without limitation device drivers, operating systems, and user applications.
  • Such computer-readable media or storage device(s) further include software for performing example aspects of the invention, as described above.
  • a module includes software, although in other example embodiments herein, a module includes hardware, or a combination of hardware and software.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Medical Informatics (AREA)
  • General Physics & Mathematics (AREA)
  • Ophthalmology & Optometry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Veterinary Medicine (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Evolutionary Computation (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Human Computer Interaction (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Pathology (AREA)
  • Psychiatry (AREA)
  • Physiology (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Eye Examination Apparatus (AREA)
  • Image Analysis (AREA)
US16/721,902 2018-12-20 2019-12-19 Detection of pathologies in ocular images Active 2040-05-28 US11503994B2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/942,132 US12042225B2 (en) 2018-12-20 2022-09-10 Detection of pathologies in ocular images
US18/748,540 US20240341589A1 (en) 2018-12-20 2024-06-20 Detection of Pathologies in Ocular Images

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
EP18214689 2018-12-20
EP18214689.4 2018-12-20
EP18214689.4A EP3671536B1 (en) 2018-12-20 2018-12-20 Detection of pathologies in ocular images

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/942,132 Continuation US12042225B2 (en) 2018-12-20 2022-09-10 Detection of pathologies in ocular images

Publications (2)

Publication Number Publication Date
US20200196853A1 US20200196853A1 (en) 2020-06-25
US11503994B2 true US11503994B2 (en) 2022-11-22

Family

ID=64755198

Family Applications (3)

Application Number Title Priority Date Filing Date
US16/721,902 Active 2040-05-28 US11503994B2 (en) 2018-12-20 2019-12-19 Detection of pathologies in ocular images
US17/942,132 Active 2040-04-29 US12042225B2 (en) 2018-12-20 2022-09-10 Detection of pathologies in ocular images
US18/748,540 Pending US20240341589A1 (en) 2018-12-20 2024-06-20 Detection of Pathologies in Ocular Images

Family Applications After (2)

Application Number Title Priority Date Filing Date
US17/942,132 Active 2040-04-29 US12042225B2 (en) 2018-12-20 2022-09-10 Detection of pathologies in ocular images
US18/748,540 Pending US20240341589A1 (en) 2018-12-20 2024-06-20 Detection of Pathologies in Ocular Images

Country Status (14)

Country Link
US (3) US11503994B2 (zh)
EP (2) EP3671536B1 (zh)
JP (4) JP6996682B2 (zh)
KR (2) KR102387820B1 (zh)
CN (2) CN118351103A (zh)
AU (2) AU2019280075B2 (zh)
CA (2) CA3064885A1 (zh)
DK (1) DK3671536T3 (zh)
ES (1) ES2980396T3 (zh)
FI (1) FI3671536T3 (zh)
MX (1) MX2021007346A (zh)
SA (1) SA521422312B1 (zh)
SG (1) SG11202106261TA (zh)
WO (1) WO2020127233A1 (zh)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12051196B2 (en) * 2018-09-12 2024-07-30 Auckland Uniservices Limited Methods and systems for ocular imaging, diagnosis and prognosis
EP3914140A4 (en) * 2019-01-25 2022-11-16 OtoNexus Medical Technologies, Inc. MACHINE LEARNING FOR OTITIS MEDIA DIAGNOSIS
EP3690888A1 (en) * 2019-01-31 2020-08-05 Nidek Co., Ltd. Ophthalmologic image processing device and ophthalmologic image processing program
JP7332463B2 (ja) * 2019-12-25 2023-08-23 キヤノン株式会社 制御装置、光干渉断層撮影装置、光干渉断層撮影装置の制御方法、及びプログラム
CN113538208B (zh) * 2020-04-20 2024-09-17 华为技术有限公司 图片加载方法及相关装置
WO2022072513A1 (en) 2020-09-30 2022-04-07 Ai-Ris LLC Retinal imaging system
WO2022158908A1 (ko) * 2021-01-22 2022-07-28 주식회사 루닛 병리 이미지 분석 방법 및 시스템
KR102379061B1 (ko) * 2021-06-30 2022-04-05 주식회사 타이로스코프 활동성 갑상선 눈병증 진료를 위한 내원 안내 방법 및 이를 수행하는 장치
JP7525851B2 (ja) 2021-06-30 2024-07-31 サイロスコープ インコーポレイテッド 活動性甲状腺眼症の医学的治療のための病院訪問ガイダンスのための方法、及びこれを実行するためのシステム
WO2023277622A1 (ko) 2021-06-30 2023-01-05 주식회사 타이로스코프 활동성 갑상선 눈병증 진료를 위한 내원 안내 방법 및 이를 수행하는 시스템
JP7521748B1 (ja) 2021-06-30 2024-07-24 サイロスコープ インコーポレイテッド 眼球突出の度合いの分析のための側方画像を取得するための方法及び撮影デバイス、及びそのための記録媒体
WO2023167102A1 (ja) 2022-03-04 2023-09-07 三菱ケミカル株式会社 炭素繊維束コンポジットの製造方法および炭素繊維束コンポジット
US20240016378A1 (en) * 2022-07-13 2024-01-18 Alcon Inc. Machine learning to assess the clinical significance of vitreous floaters

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010035607A (ja) 2008-07-31 2010-02-18 Canon Inc 診断支援装置およびその方法、プログラム、記録媒体
US20120184845A1 (en) 2010-11-11 2012-07-19 University Of Pittsburgh - Of The Commonwealth System Of Higher Education Automated macular pathology diagnosis in threedimensional (3d) spectral domain optical coherence tomography (sd-oct) images
JP2013031527A (ja) 2011-08-01 2013-02-14 Canon Inc 眼科診断支援装置および方法
US20130208966A1 (en) * 2012-02-14 2013-08-15 Tiecheng Zhao Cloud-based medical image processing system with anonymous data upload and download
US20130208955A1 (en) * 2012-02-14 2013-08-15 Tiecheng Zhao Cloud-based medical image processing system with access control
US20140276025A1 (en) * 2013-03-14 2014-09-18 Carl Zeiss Meditec, Inc. Multimodal integration of ocular data acquisition and analysis
JP2014188254A (ja) 2013-03-28 2014-10-06 Nidek Co Ltd 視野検査装置
US20160335478A1 (en) * 2014-01-28 2016-11-17 Ventana Medical Systems, Inc. Adaptive classification for whole slide tissue segmentation
US20180144470A1 (en) 2016-11-21 2018-05-24 International Business Machines Corporation Digital Data Processing for Diagnosis of a Subject
US20180140257A1 (en) 2016-11-21 2018-05-24 International Business Machines Corporation Retinal Scan Processing for Diagnosis of a Subject

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5582772B2 (ja) * 2009-12-08 2014-09-03 キヤノン株式会社 画像処理装置及び画像処理方法
JP5635898B2 (ja) * 2010-12-17 2014-12-03 キヤノン株式会社 眼底撮像装置及びその制御方法
US9955865B2 (en) * 2013-04-11 2018-05-01 Novartis Ag Method and system to detect ophthalmic tissue structure and pathologies
US8879813B1 (en) 2013-10-22 2014-11-04 Eyenuk, Inc. Systems and methods for automated interest region detection in retinal images
EP3116375B1 (en) * 2014-03-14 2023-07-19 Lkc Technologies Inc. System and method for retinopathy detection
KR102130162B1 (ko) * 2015-03-20 2020-07-06 프라운호퍼 게젤샤프트 쭈르 푀르데룽 데어 안겐반텐 포르슝 에. 베. 인공 신경망들에 대한 관련성 스코어 할당
US20170228521A1 (en) * 2016-02-05 2017-08-10 Carl Zeiss Meditec, Inc. Report driven workflow for ophthalmic image data acquisition
JP2017182739A (ja) * 2016-03-31 2017-10-05 富士通株式会社 視線検出装置、視線検出方法及び視線検出用コンピュータプログラム
JP6907563B2 (ja) 2017-01-31 2021-07-21 株式会社ニデック 画像処理装置、および画像処理プログラム
DE102017208994A1 (de) * 2017-05-29 2018-11-29 Audi Ag Verfahren zur Ermittlung von Ergebnisbilddaten

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010035607A (ja) 2008-07-31 2010-02-18 Canon Inc 診断支援装置およびその方法、プログラム、記録媒体
US20100208204A1 (en) 2008-07-31 2010-08-19 Canon Kabushiki Kaisha Eye portion diagnosis support apparatus, method therefor, program, and recording medium
US20120184845A1 (en) 2010-11-11 2012-07-19 University Of Pittsburgh - Of The Commonwealth System Of Higher Education Automated macular pathology diagnosis in threedimensional (3d) spectral domain optical coherence tomography (sd-oct) images
US20140085606A1 (en) 2011-08-01 2014-03-27 Canon Kabushiki Kaisha Ophthalmic diagnosis support apparatus and ophthalmic diagnosis support method
JP2013031527A (ja) 2011-08-01 2013-02-14 Canon Inc 眼科診断支援装置および方法
JP6025311B2 (ja) 2011-08-01 2016-11-16 キヤノン株式会社 眼科診断支援装置および方法
US20180070815A1 (en) 2011-08-01 2018-03-15 Canon Kabushiki Kaisha Ophthalmic diagnosis support apparatus and ophthalmic diagnosis support method
US20130208966A1 (en) * 2012-02-14 2013-08-15 Tiecheng Zhao Cloud-based medical image processing system with anonymous data upload and download
US20130208955A1 (en) * 2012-02-14 2013-08-15 Tiecheng Zhao Cloud-based medical image processing system with access control
US20140276025A1 (en) * 2013-03-14 2014-09-18 Carl Zeiss Meditec, Inc. Multimodal integration of ocular data acquisition and analysis
JP2016509914A (ja) 2013-03-14 2016-04-04 カール ツアイス メディテック アクチエンゲゼルシャフト 視覚データの取得および分析のマルチモード統合
JP2014188254A (ja) 2013-03-28 2014-10-06 Nidek Co Ltd 視野検査装置
US20160335478A1 (en) * 2014-01-28 2016-11-17 Ventana Medical Systems, Inc. Adaptive classification for whole slide tissue segmentation
US20180144470A1 (en) 2016-11-21 2018-05-24 International Business Machines Corporation Digital Data Processing for Diagnosis of a Subject
US20180140257A1 (en) 2016-11-21 2018-05-24 International Business Machines Corporation Retinal Scan Processing for Diagnosis of a Subject

Non-Patent Citations (14)

* Cited by examiner, † Cited by third party
Title
"ReLayNet: Retinal Layer and Fluid Segmentation of Macular Optical Coherence Tomography using Fully Convolutional Network" by Abhijit Guha Roy et al, Cornell University Library (Apr. 7, 2017), XP080761372.
ABHIJIT GUHA ROY; SAILESH CONJETI; SRI PHANI KRISHNA KARRI; DEBDOOT SHEET; AMIN KATOUZIAN; CHRISTIAN WACHINGER; NASSIR NAVAB: "ReLayNet: Retinal Layer and Fluid Segmentation of Macular Optical Coherence Tomography using Fully Convolutional Network", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 7 April 2017 (2017-04-07), 201 Olin Library Cornell University Ithaca, NY 14853 , XP080761372, DOI: 10.1364/BOE.8.003627
BURNS N, HUNTER G, JACKMAN A, HULSEY B, COUGHENOUR J, WALZ T: "The return of ozone and the hydroxyl radical to wastewater disinfection", OZONE: SCIENCE AND ENGINEERING., LEWIS PUBLISHERS, CHELSEA, MI., US, vol. 29, no. 4, 1 July 2007 (2007-07-01), US , pages 303 - 306, XP008102300, ISSN: 0191-9512, DOI: 10.1080/01919510701463206
Communication pursuant to Article 94(3) EPC issued in European Application No. 18 214 689.4-1207, (7 sheets), dated Oct. 14, 2020.
Communication pursuant to Article 94(3) EPC, dated May 3, 2022 in European Patent Application No. 18 214 689.4-1207 (7 sheets).
Examination Report dated Jul. 28, 2021 in Indian patent application No. 201944052478 (6 sheets).
Examination Report dated May 12, 2021 issued in relation to the corresponding Australian application No. 2019280075.
Extended European Search Report dated Jun. 6, 2019 relating to EP 18 214 689.4.
Jelinek H F et al"Machine learning and pattern classification in identification of indigenous retinal pathology", Engineering in Medicine and Biology Society, EMBC, 2011 33rd International Conference of the IEEE EMBS, pp. 5951-5954 (2011).
Notice of Reasons for Refusal dated Sep. 8, 2022 issued in Japanese patent application No. 2021-195247 (2 Sheets) (Machine English translation attached: 3 sheets).
Office Action dated Mar. 23, 2021, issued in Japanese Patent Application No. 2019-230618, including original Japanese version and English translation thereof.
Ruba et al: "Identification and segmentation of exudates using SVM classifier", IEEE Sponsored 2nd International Conference on Innovations in Information Embedded and Communication Systems (ICIIECS '15 )XP033192484, (6 sheets) (2015).
RUBA T.; RAMALAKSHMI K.: "Identification and segmentation of exudates using SVM classifier", 2015 INTERNATIONAL CONFERENCE ON INNOVATIONS IN INFORMATION, EMBEDDED AND COMMUNICATION SYSTEMS (ICIIECS), IEEE, 19 March 2015 (2015-03-19), pages 1 - 6, XP033192484, ISBN: 978-1-4799-6817-6, DOI: 10.1109/ICIIECS.2015.7193219
Sourya Sengupta et al., Ophthalmic Diagnosis and Deep Learning—A Survey, School of Optometry and Vision Science, University of Waterloo, Ontario, Canada, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, Dec. 9, 2018, XP08102300 (41 sheets) (available at https://www.researchgate.net/.

Also Published As

Publication number Publication date
EP4310800A3 (en) 2024-03-20
KR20200077461A (ko) 2020-06-30
MX2021007346A (es) 2021-09-30
SA521422312B1 (ar) 2024-01-08
AU2021232682A9 (en) 2021-12-02
EP3671536B1 (en) 2024-05-22
US20200196853A1 (en) 2020-06-25
CN118351103A (zh) 2024-07-16
KR102494503B1 (ko) 2023-01-31
WO2020127233A1 (en) 2020-06-25
JP7279291B2 (ja) 2023-05-23
ES2980396T3 (es) 2024-10-01
US20230000338A1 (en) 2023-01-05
AU2019280075A1 (en) 2020-07-09
JP7509505B2 (ja) 2024-07-02
EP4310800A2 (en) 2024-01-24
JP6996682B2 (ja) 2022-01-17
JP2023021440A (ja) 2023-02-10
KR20220027927A (ko) 2022-03-08
CA3223918A1 (en) 2020-06-20
AU2021232682B2 (en) 2022-11-24
CN111353970B (zh) 2024-05-14
US12042225B2 (en) 2024-07-23
CN111353970A (zh) 2020-06-30
AU2019280075B2 (en) 2021-07-22
DK3671536T3 (da) 2024-06-17
JP2020124472A (ja) 2020-08-20
CA3064885A1 (en) 2020-06-20
FI3671536T3 (fi) 2024-06-18
JP2022033126A (ja) 2022-02-28
AU2021232682A1 (en) 2021-10-07
JP2024119998A (ja) 2024-09-03
EP3671536A1 (en) 2020-06-24
KR102387820B1 (ko) 2022-04-15
SG11202106261TA (en) 2021-07-29
US20240341589A1 (en) 2024-10-17

Similar Documents

Publication Publication Date Title
US12042225B2 (en) Detection of pathologies in ocular images
US12040079B2 (en) Medical image processing apparatus, medical image processing method and computer-readable medium
JP7229881B2 (ja) 医用画像処理装置、学習済モデル、医用画像処理方法及びプログラム
JP7269413B2 (ja) 医用画像処理装置、医用画像処理システム、医用画像処理方法及びプログラム
CN105310645B (zh) 图像处理装置及图像处理方法
US11922601B2 (en) Medical image processing apparatus, medical image processing method and computer-readable medium
US11284791B2 (en) Image processing method, program, and image processing device
JP6878923B2 (ja) 画像処理装置、画像処理システム、および画像処理プログラム
CN113226153A (zh) 图像处理装置、图像处理方法和程序
EP3719808B1 (en) Determining levels of hypertension from retinal vasculature images
Sánchez et al. Improving hard exudate detection in retinal images through a combination of local and contextual information
WO2022157838A1 (ja) 画像処理方法、プログラム、画像処理装置、眼科システム
JP2021154159A (ja) 機械学習ガイド付き撮影システム
TR2021007756A2 (tr) Reti̇na görüntü anotasyonu ve i̇lgi̇li̇ eği̇ti̇m yöntemleri̇ i̇le görüntü i̇şleme modelleri̇

Legal Events

Date Code Title Description
FEPP Fee payment procedure

Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STPP Information on status: patent application and granting procedure in general

Free format text: SENT TO CLASSIFICATION CONTRACTOR

AS Assignment

Owner name: OPTOS PLC, GREAT BRITAIN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:VAN HEMERT, JANO;SWAN, DEREK;SIGNING DATES FROM 20191210 TO 20200109;REEL/FRAME:052993/0422

STPP Information on status: patent application and granting procedure in general

Free format text: APPLICATION DISPATCHED FROM PREEXAM, NOT YET DOCKETED

STPP Information on status: patent application and granting procedure in general

Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION

STPP Information on status: patent application and granting procedure in general

Free format text: NON FINAL ACTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER

STPP Information on status: patent application and granting procedure in general

Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS

STPP Information on status: patent application and granting procedure in general

Free format text: AWAITING TC RESP., ISSUE FEE NOT PAID

STPP Information on status: patent application and granting procedure in general

Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS

STPP Information on status: patent application and granting procedure in general

Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED

STCF Information on status: patent grant

Free format text: PATENTED CASE

STPP Information on status: patent application and granting procedure in general

Free format text: AWAITING TC RESP., ISSUE FEE NOT PAID

STCF Information on status: patent grant

Free format text: PATENTED CASE